Top Banner
BioMed Central Page 1 of 14 (page number not for citation purposes) Malaria Journal Open Access Research A rapid malaria appraisal in the Venezuelan Amazon Wolfram G Metzger 1,2,4 , Anibal M Giron 2 , Sarai Vivas-Martínez 1,3 , Julio González 2 , Antonio J Charrasco 2 , Benjamin G Mordmüller 4 and Magda Magris* 1 Address: 1 Servicio Autónomo Centro Amazónico para la Investigación y Control de Enfermedades Tropicales, Simón Bolívar' (SACAICET), Puerto Ayacucho, Estado Amazonas, Venezuela, 2 Dirección Regional de Salud Amazonas, Ministerio del Poder Popular para la Salud, Puerto Ayacucho, Estado Amazonas, Venezuela, 3 Cátedra de Salud Pública. Escuela de Medicina Luis Razetti. Universidad Central de Venezuela, Caracas, Venezuela and 4 Eberhard Karls Universität Tübingen, Institut für Tropenmedizin, Tübingen, Germany Email: Wolfram G Metzger - [email protected]; Anibal M Giron - [email protected]; Sarai Vivas-Martínez - [email protected]; Julio González - [email protected]; Antonio J Charrasco - [email protected]; Benjamin G Mordmüller - [email protected]; Magda Magris* - [email protected] * Corresponding author Abstract Background: While the federal state of Amazonas bears the highest risk for malaria in Venezuela (2007: 68.4 cases/1000 inhabitants), little comprehensive information about the malaria situation is available from this area. The purpose of this rapid malaria appraisal (RMA) was to provide baseline data about malaria and malaria control in Amazonas. Methods: The RMA methodology corresponds to a rapid health impact assessment (HIA) as described in the 1999 Gothenburg consensus. In conjunction with the actors of the malaria surveillance system, all useful data and information, which were accessible within a limited time- frame of five visits to Amazonas, were collected, analysed and interpreted. Results: Mortality from malaria is low (< 1 in 10 5 ) and slide positivity rates have stayed at the same level for the last two decades (15% ± 6% (SD)). Active case detection accounts for ca. 40% of slides taken. The coverage of the censured population with malaria notification points (NPs) has been achieved in recent years. The main parasite is Plasmodium vivax (84% of cases). The proportion of Plasmodium falciparum is on the decline, possibly driven by the introduction of cost-free artemisinin- based combination therapy (ACT) (1988: 33.4%; 2007: 15.4%). Monitoring and documentation is complete, systematic and consistent, but poorly digitalized. Malaria transmission displayed a visible lag behind rainfall in the capital municipality of Atures, but not in the other municipalities. In comparison to reference microscopy, quality of field microscopy and rapid diagnostic tests (RDTs) is suboptimal (kappa < 0.75). Hot spots of malaria risk were seen in some indigenous ethnic groups. Conflicting strategies in respect of training of community health workers (CHW) and the introduction of new diagnostic tools (RDTs) were observed. Conclusion: Malaria control is possible, even in tropical rain forest areas, if the health system is working adequately. Interventions have to be carefully designed and the features of the particular local Latin American context considered. Published: 11 December 2009 Malaria Journal 2009, 8:291 doi:10.1186/1475-2875-8-291 Received: 22 June 2009 Accepted: 11 December 2009 This article is available from: http://www.malariajournal.com/content/8/1/291 © 2009 Metzger et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0 ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
14

A rapid malaria appraisal in the Venezuelan Amazon

Dec 19, 2022

Download

Documents

Heinrich Härke
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: A rapid malaria appraisal in the Venezuelan Amazon

BioMed CentralMalaria Journal

ss

Open AcceResearchA rapid malaria appraisal in the Venezuelan AmazonWolfram G Metzger1,2,4, Anibal M Giron2, Sarai Vivas-Martínez1,3, Julio González2, Antonio J Charrasco2, Benjamin G Mordmüller4 and Magda Magris*1

Address: 1Servicio Autónomo Centro Amazónico para la Investigación y Control de Enfermedades Tropicales, Simón Bolívar' (SACAICET), Puerto Ayacucho, Estado Amazonas, Venezuela, 2Dirección Regional de Salud Amazonas, Ministerio del Poder Popular para la Salud, Puerto Ayacucho, Estado Amazonas, Venezuela, 3Cátedra de Salud Pública. Escuela de Medicina Luis Razetti. Universidad Central de Venezuela, Caracas, Venezuela and 4Eberhard Karls Universität Tübingen, Institut für Tropenmedizin, Tübingen, Germany

Email: Wolfram G Metzger - [email protected]; Anibal M Giron - [email protected]; Sarai Vivas-Martínez - [email protected]; Julio González - [email protected]; Antonio J Charrasco - [email protected]; Benjamin G Mordmüller - [email protected]; Magda Magris* - [email protected]

* Corresponding author

AbstractBackground: While the federal state of Amazonas bears the highest risk for malaria in Venezuela(2007: 68.4 cases/1000 inhabitants), little comprehensive information about the malaria situation isavailable from this area. The purpose of this rapid malaria appraisal (RMA) was to provide baselinedata about malaria and malaria control in Amazonas.

Methods: The RMA methodology corresponds to a rapid health impact assessment (HIA) asdescribed in the 1999 Gothenburg consensus. In conjunction with the actors of the malariasurveillance system, all useful data and information, which were accessible within a limited time-frame of five visits to Amazonas, were collected, analysed and interpreted.

Results: Mortality from malaria is low (< 1 in 105) and slide positivity rates have stayed at the samelevel for the last two decades (15% ± 6% (SD)). Active case detection accounts for ca. 40% of slidestaken. The coverage of the censured population with malaria notification points (NPs) has beenachieved in recent years. The main parasite is Plasmodium vivax (84% of cases). The proportion ofPlasmodium falciparum is on the decline, possibly driven by the introduction of cost-free artemisinin-based combination therapy (ACT) (1988: 33.4%; 2007: 15.4%). Monitoring and documentation iscomplete, systematic and consistent, but poorly digitalized. Malaria transmission displayed a visiblelag behind rainfall in the capital municipality of Atures, but not in the other municipalities. Incomparison to reference microscopy, quality of field microscopy and rapid diagnostic tests (RDTs)is suboptimal (kappa < 0.75). Hot spots of malaria risk were seen in some indigenous ethnic groups.Conflicting strategies in respect of training of community health workers (CHW) and theintroduction of new diagnostic tools (RDTs) were observed.

Conclusion: Malaria control is possible, even in tropical rain forest areas, if the health system isworking adequately. Interventions have to be carefully designed and the features of the particularlocal Latin American context considered.

Published: 11 December 2009

Malaria Journal 2009, 8:291 doi:10.1186/1475-2875-8-291

Received: 22 June 2009Accepted: 11 December 2009

This article is available from: http://www.malariajournal.com/content/8/1/291

© 2009 Metzger et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Page 1 of 14(page number not for citation purposes)

Page 2: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

BackgroundRoutine malaria surveillance systems, and the data theyproduce, need to be analysed, otherwise many useful datamay become lost. Yet it is important to note that, outsideof experimental trials, the effectiveness of interventionssuch as, for example, insecticide-treated bed nets, can onlybe evaluated when routine baseline data has already beenassessed. In 2007, a rapid malaria appraisal (RMA) wascarried out as part of a project providing technical assist-ance to the malaria surveillance system in the Federal Stateof Amazonas, Venezuela. The purpose of this project wasto provide baseline data and facilitate operationalchanges. In what follows, results of this evaluation are pre-sented in brief.

The Federal State of Amazonas (180,145 km2) covers anarea as big as the UK, excluding Scotland. More than halfof the 135,585 inhabitants live in the capital PuertoAyacucho, which is situated in the far North-West of theterritory. Amazonas is the most southern federal state ofVenezuela, bordering Colombia to the west and Brazil tothe east. It comprises seven municipalities characterizedby very low density of population, remoteness of commu-nities, absence of roads, and transport via river or air. Halfof the population originates from 19 indigenous ethnicgroups. In the municipality of Alto Orinoco live about12,000 Yanomami people, 5,000 of whom it is estimatedcontinue to reside in the vast expanse of the tropical forestremaining uncaptured by any census [1]. Unlike theneighbouring state of Bolivar, Amazonas does not hostsignificant gold mining activities.

The malaria surveillance system in Venezuela wasfounded in 1936 by Arnoldo Gabaldón, who became alegendary figure in the history of public health in Vene-zuela. For a long time this institution (known as malario-logia) acted with a high degree of autonomy. Based onDDT spraying, early diagnosis and therapy, it wasdesigned as a vertical programme with teams of healthworkers visiting the communities. The drug distributionservice was probably the first of its type to make anti-malarials available, without payment, to rich and pooralike [2]. Partly as result of these activities, in 1961 WHOdeclared the eradication of malaria in the major part (2/3)of the then malarious areas of Venezuela [3]. However,the Amazon region was excluded from eradication effortsbecause malaria there was designated as inaccessible.

Malaria research activities in Amazonas have been infre-quent, but have increased in recent years. A search forpeer-reviewed articles containing data on clinical malaria,mainly from cross-sectional studies, listed 18 publicationsfrom 1990 onward [4-21]. The majority (12/18) were car-ried out in the Yanomami people of the southern munic-ipality Alto Orinoco. Very few data were from Atures, even

though most malaria cases are from this northern munic-ipality (see also Malaria Atlas Project 2007 [22]). Entomo-logical data are presented in various publicationsdescribing Anopheles darlingi as the primary vector [17].

In 2002, the malaria surveillance system was integrated bythe Ministry of Health into the main epidemiological sur-veillance system. As for supporting infrastructure, to date,the governmental health sector of Amazonas counts onehospital constructed in the 1950s (100 beds), one mod-ern Popular Clinic, six Comprehensive Health Centres inconstruction, six Rehabilitation Centers, seven urban and92 rural health posts, as well as 32 health posts of the Bar-rio Adentro programme [23,24]. A modern hospital (220beds) is under construction and will be completed in2010. The Venezuelan constitution (1999) guaranteescost-free medical care and forbids the privatization of thepublic health system.

MethodsThis work was done within the framework of providingtechnical assistance to the malaria surveillance system ofthe federal state of Amazonas, Venezuela, carried outbetween October 2007 and August 2008. The study repre-sented a rapid malaria appraisal (RMA) corresponding toa rapid health impact appraisal (HIA) under resource-con-strained conditions [25]. Established guidelines for mon-itoring and evaluation [26] were followed and modifiedaccording to local needs. The study comprised retrospec-tive, concurrent and prospective aspects.

Five visits were carried out. The duration of each wasbetween five days and five weeks. Field investigationswere undertaken in the capital municipality of Atures, andshort visits were carried out to the municipalities ofAutana, Atabapo, Manapiare, and Alto Orinoco. The docu-mentation system was analysed, data were extracted andadditional data were collected. Structured and semi-struc-tured questionnaires were applied with individual andgroup interviews being conducted. A pilot study to assessthe quality of malaria diagnoses (microscopy, Rapid Diag-nostic Tests) was carried out and workshops were held atthe end of each visit. All statistical analyses were doneusing R v2.7 [27]. Time series were analysed for trend andserial dependence.

ResultsMalaria in Venezuela and Amazonas, 1988 - 2007In Venezuela, the number of malaria cases increasedwithin two decades from 22,056 in 1988 to 41,570 in2007 displaying a peak in 2004 (46,244 cases). Whilstroughly 20% of malaria cases were due to Plasmodium fal-ciparum, overwhelmingly the rest (nearly 80%) wereattributed to Plasmodium vivax. Interestingly, Plasmodiummalariae and mixed infections are only marginally

Page 2 of 14(page number not for citation purposes)

Page 3: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

detected. Mortality from malaria (using ICD-9 (084) andICD-10 (B50-B54) is low and fell significantly from 45(1988) to 16 (2007) fatal cases. A linear model of the datashowed a decrease of 1.47 deaths per year (Figure 1) (p =0.01) [28]. In 2007, Venezuela counted with a populationof 26,02 × 106million inhabitants [29] resulting in acountrywide mortality rate from malaria of < 1 in 106

inhabitants.

Amazonas contributes about a fifth (2007: 22%) of thetotal of malaria cases in Venezuela and ranks second, inregard to absolute case numbers, behind Bolivar. How-ever, due to its low population density, it displays thehighest incidence per head of population (2007: 68.4cases/1,000 inhabitants). The analysis of mortalityrecords shows that mortality due to malaria in Amazonasis low and has remained at the same level for the lasttwenty years (p = 0.68) (Figure 1) [28].

In the same period, the number of malaria cases detectedin Amazonas multiplied nearly nine times (1988: 1,297cases; 2007: 9,204 cases). Within this period, the popula-tion almost doubled (1988: 75,830 inhabitants; 2007:134,585 inhabitants) and the number of slides takenincreased more than eight times (1988: 10,736 slides;2007: 88,079). Also the coverage of permanent activemalaria notification points (NPs) rose from 2 NPs (1988)

to 82 NPs (2007). The majority of NPs were establishedbetween 2003 and 2005, which is reflected in the increasein slides taken within that period (Figure 2).

Consequently, the annual parasite index (API, malariacases per 1,000 persons per year) for malaria casesincreased from 17 (1988) to 68 (2007) (p = 0.021), butthe slide positivity rate (SPR, malaria cases per 100 slidestaken) did not change significantly for the last 20 years(SPR 1988: 12; SPR 2007: 10) (p = 0.17) (Figure 3).

Subgrouping for malaria species revealed that the percent-age of P. falciparum decreased steadily (1988: 33.4%;2007: 15.4%) which is reflected in different patterns ofthe SPR for the two species. The SPR of P. vivax remainedon the same level (p = 0.43), whereas the SPR of P. falci-parum decreased significantly (p = 0.018). (Figure 4).

Rainfall and malaria transmission in Amazonas municipalities, 1998 - 2007Subanalysis for the municipalities of Amazonas could becarried out for the decade 1998 to 2007 only, becauseprior to this period a reform at local government level hadbeen undertaken, changing names and territory of someof the municipalities. Malaria transmission displayed avisible lag behind peaks of rainfall in the capital munici-pality of Atures (Figure 5, 6), which was not seen in the

Malaria Mortality in Venezuela and Amazonas (1988-2007)Figure 1Malaria Mortality in Venezuela and Amazonas (1988-2007). Fatal malaria cases in Venezuela (bright red), fatal malaria cases in Amazonas (dark red).

30

40

50

60

70

Mal

aria

Cas

es

0

10

20

30

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Fat

al M

Years

Page 3 of 14(page number not for citation purposes)

Page 4: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

other municipalities. Partial autocorrelation of datasetsshowed a significant positive correlation of rainfall withthe incidence of P. vivax with a lag of 4 months. This pat-tern was even more pronounced in P. falciparum whereparasite incidence was significantly lagging two to fivemonths behind rainfall. Negative correlations of P. falci-parum, that fit the expected seasonal pattern, could be seen1, 7, 10, and 11 months after rainfall.

Malaria species, sex, age and ethnic differences, 2007In 2007, out of the 9,204 cases found positive for malariain Amazonas, 447 (4.8%) were imported, mainly fromColombia. Out of the autochthonous cases (n = 8757),more than half of the patients came from the municipalityof Atures, the rest were distributed among the othermunicipalities (Table 1) [30].

7736 cases (83%) were caused by P. vivax, 1416 (16%)cases were caused by P. falciparum, 12 cases (0.1%) caseswere attributed to P. malariae, and 40 cases (0.4%) weremixed infections. Interestingly, the distribution of speciesshowed significant differences between ethnic groups. Forexample, the Yanomami group displayed a significantlyhigher rate of P. falciparum infections than the other eth-nic groups (Yanomami 40.3%, Yekwana 22.4%, Puinabe

16.2%, Guahibo 14.9%, Baré 12.1%, Criollo 11.2%, Cur-ripaco 10.9%, Piapaco 9.2%, Piaroa 8.7%). Subgroupingby age groups showed that 17,2% of cases were in patientsunder five years of age (Figure 7).

The overall risk of malaria showed considerable differ-ences between the municipalities. This picture changednoticeably when ethnic groups were mapped and a higherrisk was revealed for some ethnic groups in comparison tothe neighbouring population (Figure 8).

Structure of the malaria surveillance system in AmazonasThe malaria surveillance system in Amazonas employsaround 150 persons, half of them with renewable short-term contracts. It is embedded in the Service of Environ-mental Surveillance which runs also the control of dengueand intestinal parasites. Malaria Control is headed by acoordinator and divided into sections of Vector Control,Medical Attention and Entomology. The main actors areinspectors (two years training in the central school formalaria inspectors in Maracay, Venezuela, microscopists(three to six months training as auxiliary nurses), visitorsand fumigators (two months training as auxiliary staff).The distribution of malaria personnel reflects roughly thepopulation density of the municipalities (Table 1) [30].

Malaria Cases, Slides taken, Population in Amazonas (1988-2007)Figure 2Malaria Cases, Slides taken, Population in Amazonas (1988-2007). Malaria cases (red, right vertical axis), slides taken (blue, left vertical axist), population (grey, left vertical axis).

6000

8000

10000

12000

14000

60000

80000

100000

120000

140000

lari

a C

ases

ula

tio

n, S

lides

0

2000

4000

6000

0

20000

40000

60000

19881989199019911992199319941995199619971998199920002001200220032004200520062007

Ma

Po

pu

Years

Page 4 of 14(page number not for citation purposes)

Page 5: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

Aside from the traditional system, the 'Andean Plan forMalaria in Frontier Areas' (PAMAFRO, Plan Andino deMalaria en Areas Fronterizas) has been operating in Vene-zuela since 2005. This non-governmental organization isfinanced by the Global Fund for the Fight against AIDS,Tuberculosis and Malaria. One focus of its strategy is theparticipation of the communities on a voluntary basis[31]. To date, about 150 community health workers havebeen trained within this programme.

The states surveillance system and PAMAFRO, operatedquasi-independently of each other until 2008, whenPAMAFRO was integrated into the central coordination ofthe malaria surveillance system.

Knowledge of malaria staff about malariaAn anonymous self-made questionnaire was given out toexplore the knowledge about malaria of the personnel(technical and non-technical) working in the control sys-tem of the municipality of Atures. Answers to six multiplechoice questions about malaria were requested (see addi-tional file 1). Out of 75 employees, 52 returned theirsheets fully completed. Results were, in brief: 1) 92% (48/52) knew the transmission cycle of Plasmodium. 2) 54%

(28/52) could identify the three types of malaria in Vene-zuela. 3) 73% (38/52) identified P. falciparum as the mostdangerous of the parasites. 4) 46% (42/52) knew that allthree medicaments chloroquine, artesunate, and meflo-quine were anti-malarials. 6% (3/52) knew that also dox-ycycline had anti-malarial activity. 5) 81% (42/52) wereaware that the duration of the treatment can be dependenton the type of the parasite, and 6) 48% (25/52) knewthree methods for preventing malaria (e.g. impregnatedmosquito nets, fumigation/indoor spraying, and avoidingbreeding sites).

Educational level and type of work were correlated toknowledge about malaria. When stratified for academiclevel, 20% (3/15) of unskilled workers, 67% (14/21) ofemployees with A-level (advanced training after second-ary school), and 69% (11/16) of technical staff, respec-tively, identified the three main malaria parasites presentin Venezuela. When stratified for medical or non-medicalemployees (inspectors/microscopists/visitors vs. fumiga-tors/drivers/secretaries), 84% (21/25) of those working inmedical activities, and 44% (12/27) of those working innon-medical activities, respectively, knew that artesunateis an anti-malarial drug.

Annual Parasite Index and Slide Positivity Rate in Amazonas (1988 - 2007)Figure 3Annual Parasite Index and Slide Positivity Rate in Amazonas (1988 - 2007). Annual parasite index (grey, API), slide positivity rate (blue, SPR).

50,00

60,00

70,00

80,00

90,00

100,00A

PI,

SP

R

0,00

10,00

20,00

30,00

40,00

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

A

Years

Page 5 of 14(page number not for citation purposes)

Page 6: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

Treatment and adherenceMalaria treatment in the Venezuelan Amazon is exclu-sively in the hands of the malaria surveillance system. Nomalaria medicaments are sold in pharmacies and there areno other sources of anti-malarial drugs. Little is knownabout the impact of drugs crossing the border from neigh-bouring Colombia or Brazil. However, as the publichealth system of Venezuela provides cost-free treatment,in any event, it can be assumed that traffic of medicamentsis rather from Venezuela to neighbouring countries thanvice versa. In 2004, the Venezuelan Health Ministry estab-lished artemisinin-based combination therapy (ACT) asfirst-line treatment for P. falciparum (Artesunat®, Meflo-quine®, and primaquine). For the treatment of P. vivax,chloroquine and primaquine is applied. ACT had a visibleimpact on transmission of P. falciparum (Figure 5). Theimpact on P. vivax was less pronounced (Figure 6).

The guidelines of the malaria control programme implythat treatment is taken under the supervision of healthpersonnel. However, due to long distances and socio-cul-tural conditions, this is often impossible. Thus, little isknown about adherence to malaria treatment and itsimpact on the development of resistance due to sub-ther-apeutic dosages by incomplete treatment regimens.

Detection of casesMalaria surveillance in Amazonas is based on the earlydiagnosis of cases through active and passive case detec-tion. Passive detection is realized through 82 NPs, locatedin health posts, hospitals, consultancies and missionposts. Most of them are run by microscopists, who alsoadminister treatment. All medical professionals, includ-ing private doctors, are obliged to send suspected malariacases to the nearest NP. Active case detection is realized by24 visitors. These health workers are sent to areas andcommunities with high incidence of malaria or where epi-demics are anticipated. Besides active and passive casedetection, 150 voluntary community health workerstrained by PAMAFRO (see above) contribute to the detec-tion of malaria cases.

The proportion of slides taken by active and passive casedetection did not change significantly during twentyyears: in 1988, 40% (4330/10736) of slides were taken byactive case detection, and in 2007, 41% (36112/80969) ofslides were taken by active case detection. Activelydetected cases did not increase with the increase in theabsolute number of actively taken slides (r = 0.24, p =0.29) (Figure 9), However, the increase of passively takenslides correlated significantly with the number of pas-

Slide Positivity Rate of P. falciparum and P. vivax in Amazonas (1988 - 2007)Figure 4Slide Positivity Rate of P. falciparum and P. vivax in Amazonas (1988 - 2007). Slide positivity rate (SPR) of P. vivax (bright blue), slide positivity rate (SPR) of P. falciparum (dark blue).

15

20

25S

PR

0

5

10

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Years

Page 6 of 14(page number not for citation purposes)

Page 7: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

sively detected cases (r = 0.87, p < 0.001) (Figure 10). In2007, only 27 cases (0.3%) were registered as casesdetected by voluntary community health workers trainedby PAMAFRO.

Quality assessment of malaria diagnosisMicroscopyIn 2007, 88 079 microscopic slides were examined inAmazonas (Figure 2). The malaria surveillance systemstipulates a double quality check: All positive slides and10% of negative slides are to be re-examined by micro-scopists in the Central Malaria Laboratory (CML) inPuerto Ayacucho. Then, these slides are sent to theNational Malaria Reference Laboratory in Maracay, Vene-zuela, for a second re-examination. The concordance ofslide reading between health posts and CML was reportedto be 99%. However, the slides in the CML were re-exam-ined in a non-blinded manner. A pilot study was carriedout in order to read 1,000 slides in a blinded way. Resultsshowed a lower concordance (Cohens Kappa Index =0.74) and considerable differences between particularmicroscopists in the health posts due to different levels ofrefresher training.

Rapid Diagnostic Tests (RDTs)In 2007, nearly 7,000 RDTs were given out by PAMAFROto microscopists and health workers for use and evalua-

tion. However, it was observed that many microscopistsceased microscopy and used RDTs only. This was in starkcontrast to the low confidence of microscopists in RDTs,because it was known that the performance of RDTs is notoptimal when the parasitaemia is low and storage condi-tions are inadequate (as is the case in most areas of Ama-zonas). In order to improve quality assessment andconfidence in the adequate routine use of RDTs, a pilotstudy with the systematic evaluation of 500 RDTs was per-formed. Results showed that concordance with micros-copy was good, but not optimal (Cohens Kappa Index =0.72) (manuscript in preparation).

Documentation systemThe flow of information and the documentation of activ-ities is mainly handled with paper and pencil by the localmalaria teams. Many records are kept as in ArnoldoGabaldón's times (see above). Digitalization has beenintroduced at some places for practical reasons, but amore systemic approach is required. Thus, many data ofhigh quality, for example data for temporal and spatialtransmission (Figures 5 and 6), are saved impeccably, butthey are not easily accessible to modern research methods.In workshops it has been agreed that a great number ofsuperfluous formats could be eliminated, and others havebeen transformed into a single format format. Training

P. vivax Slide Positivity Rate and Rainfall in Atures/Amazonas (1998 - 2007)Figure 5P. vivax Slide Positivity Rate and Rainfall in Atures/Amazonas (1998 - 2007). Slide positivity rate P. vivax (blue, SPR), rainfall (green).

20,00

25,00

30,00

35,00

40,00

45,00

500

600

700

800

900

1000

SP

R P

. viv

ax

Rai

nfa

ll

Artemisinine Combination Therapy

(ACT)

0,00

5,00

10,00

15,00

0

100

200

300

400

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

S

Years

Page 7 of 14(page number not for citation purposes)

Page 8: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

courses for staff members in data management andprocessing have been planned and implemented.

An important factor concerning the flow of information isthe limited infrastructure. When transport is possible onlyvia river or air, news from far away health posts dependson weather conditions and available logistics. Normally,

written data and slides arrive within a week in the capitalPuerto Ayacucho. As malaria cases in far away health postsare also reported via radio communication, another epi-demiological data base is created using this route. Thisdata base sometimes includes data slightly different fromthe malaria surveillance documentation, since the latter,for example, includes only double-checked cases.

P. falc. Slide Positivity Rate and Rainfall in Atures/Amazonas (1998 - 2007)Figure 6P. falc. Slide Positivity Rate and Rainfall in Atures/Amazonas (1998 - 2007). Slide positivity rate P. falciparum (blue, SPR), rainfall (green).

8,00

10,00

12,00

14,00

16,00

500

600

700

800

900

1000

R P

. fal

cip

aru

m

Rai

nfa

ll

Artemisinine Combination Therapy

(ACT)

0,00

2,00

4,00

6,00

0

100

200

300

400

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

SP

R

Years

Table 1: Malaria and malaria control in Amazonas

Population Malaria Cases NP* Microscopists Visitors Inspectors

Atures 86,368 4,342 32 13 10 4

Autana 7,693 1,339 12 7 3 1

Atabapo 9,991 1,005 6 3 2 1

Manapiare 9,063 813 19 10 2 1

Alto Orinoco 17,058 1,165 11 3 5 1

Maroa 1,848 5 1 1 0 0

Rio Negro 2,564 88 1 1 0 0

* Notification Point

Page 8 of 14(page number not for citation purposes)

Page 9: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

Entomology and vector controlEntomological activities comprised supervision of breed-ing sites including investigations for the evaluation of bio-logical larvicides (Bacillus sphaericus). Breeding sites weremapped on paper formats. Night catches are carried outthroughout the year, but are limited by logistic con-straints. In 2007, 1,016 Anopheles darlingi (2006: 3252)and 574 Anopheles braziliensis (2006: 1622) were caughtand identified. Roughly half of the mosquitos were cap-tured indoors. Peaks of night catches coincided with peaksof rain during the months of September, October, andNovember.

Vector control activities comprised indoor residual spray-ing and fumigation. Sprayings and fumigations were car-ried out in endemic areas or where epidemics wereanticipated. Anticipation of epidemics is calculated byinspectors through the analysis of local malaria epidemi-ology in the respective communities. In 2007, 5,288houses were sprayed (2006: 3,350) and and the fumiga-tion of 339,009 houses was performed (2006: 396 348).The insecticides used were fenothion powder (2007:7,107 kg), fenothion emulsion (2007: 10,920 litres), del-tamethrin (2007: 9,153.7 litres) for spraying of houses.For fumigation, malathion (2007: 18,078.32 litres), andfenothion (2007: 11,187 litres) were used.

The population of Amazonas is familiar with ITNs, but thecoverage of effective bed net usage is rarely documented[14]. In 2007, 10,000 impregnated mosquito nets weredistributed in the municipality of Atures, mainly in indig-enous communities. The analysis of its impact will bedescribed elsewhere.

DiscussionHealth care for malaria patients in Amazonas is guaran-teed and mortality from malaria is low. However, malaria,as dengue fever, is a sensitive issue in Venezuela. The riseand fall in malaria cases serves as a topic for the politics ofthe day. In this context, it is important to show thatmalaria epidemiology is dependent on many factors andthorough analysis is needed to describe these trends andexplain the reasons. A first glance at the malaria surveil-lance data of Amazonas shows a rising number of malariacases during the last two decades and, even when the risein population is considered, the Annual Parasite Index(API) is increasing. However, at the same time the numberof notification points (NPs) increased as well, and thepositivity rate of slides (SPR) maintained a steady state forthe last two decades.

Moreover, subanalysis of species reveals that the propor-tion of P. falciparum cases actually decreased, whereas P.

Malaria Infections by Age Groups in Amazonas (2007)Figure 7Malaria Infections by Age Groups in Amazonas (2007). Malaria cases of P. falciparum (dark red), P. vivax (bright red), P. malariae (purple), mixed infections (brown).

100

1000

10000la

ria

Cas

es

1

10

< 5 6-10 11-20 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 56-60 61-65 66-70 71-75 75-80 > 80

Ma

Age (Years)

Page 9 of 14(page number not for citation purposes)

Page 10: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

vivax stayed at the same level. Thus, host immunity,malaria surveillance and parasite transmission has kept abalance over the years and the existing risk of gettingmalaria has been accompanied by low mortality. Inrespect of P. falciparum, this equilibrium was altered posi-tively by the introduction of ACT.

In 2003, the malaria surveillance system in the state ofSucre - the third federal state of Venezuela endemic formalaria, behind Amazonas and Bolivar - reported successreducing the API from 20,1 to 6,1. This was mainly theresult of Roll Back Malaria interventions. Temporal andspatial transmission patterns had been analysed and masstreatment and integrated vector control measures, accom-panied by health education, community participationand an overall strentgthening of the health services, hadbeen carried out [32-35].

A key issue in public health is equity, which can be estab-lished through access to diagnosis and treatment. Ensur-ing that a high proportion of patients have access toeffective treatment will be essential to malaria eliminationprogrammes [36]. In Amazonas, the traditional malariasurveillance system is backed by the government's deci-sion to provide free health care for all. It can be stated that

- with the exception of the part of the Yanomami popula-tion living reclusively in remote areas of the Alto OrinocoCasiquiare Biosphere Reserve - free access to malaria diag-nosis and treatment is provided for the entire population.

Apart from the biosphere reserve, the Venezuelan Amazondisplays two zones of malaria transmission: the moredensely populated northern part shows a seasonal patternwith peaks of malaria cases lagging three to four monthsbehind rainfall peaks. This area is characterized by asavanna-like geography and easy access to health posts.Communities in the less densely populated southern part,live alongside rivers at the edge of the rainforest. Patientsoften have to reach a health post by foot or by boat. Noseasonal pattern of malaria transmission could bedetected here. This lack of seasonality was also noticed inan earlier epidemiological study in Alto Orinoco [17].

One important finding was that ethnic groups display dif-ferent risks for malaria. This might be the consequence ofdiffering socio-economic conditions, living habits andpatterns of adherence to malaria drugs. For example, thecriollo (mixed) population living in the urbanized area ofPuerto Ayacucho showed a much lower risk for malariathan the indigenous Guahibo living in rural areas around

Annual Parasite Index (API) by Municipalities and by Ethnic Groups in Amazonas (2007)Figure 8Annual Parasite Index (API) by Municipalities and by Ethnic Groups in Amazonas (2007). Left: Annual Parasite Index (API) by municipalities. Right: Annual Parasite Index (API) by ethnic groups. Schematic illustration of the settlement area of the ethnic groups of Guahibo (1), Puinave (2), Yanomami (3).

51 – 100

101 - 150

151 - 200

201 - 250

API:

0 – 50

1

2

3

Page 10 of 14(page number not for citation purposes)

Page 11: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

the city. Future interventions should focus on these "hotspots" of transmission. As a consequence of this, aresearch project for the evaluation of blister packed treat-ment was implemented in 2008, which considers age andweight as well as the ethnic background of the patients.

The purpose of this RMA and provision of technical assist-ance was to provide baseline data for possible interven-tions as well as to define and facilitate reorganization, if itis necessary. In initial interviews with key players it wasagreed that the focus of this project should be the evalua-tion of the surveillance data and the assessment of malariadiagnosis. Cross-sectional school parasitaemia andhealth-facility based surveys - as proposed in the rapidurban malaria appraisal (RUMA) methodology of Africancities [37] - were not prioritized in this RMA, because themonitoring system in Amazonas functions adequatelyand the quality of the routine health statistics was foundto be high. Thus, it could be assumed that all suspectedcases were captured and confirmed microscopically.

It is appropriate to discuss and plan the final design of aRMA and technical assistance project in a horizontal man-ner with the actors involved. As conditions differ locallythe methodology has to be adapted accordingly [38].

Fixed protocols appear advantageous at first glance, butsome elements might turn out to be unfeasible underresource-constrained conditions. For example, mappingof breeding sites might be possible only under certainconditions [39-42]. Hence, in this RMA it was decidedthat the entomological and vector control unit would besubject to preliminary evaluation only, to be followed bya separate evaluation project.

Small steps - appearing unimpressive at first sight - werecrucial for later success. For example, before carrying outthe blinded double-check of field slides, microscopistshad to be convinced that conflicting results from tworeaders are nothing to be ashamed of. Also, the digitaliza-tion of a hand-written documentation system, which hadstood the test of time for half a century, had to be dis-cussed in detail with the personnel involved and requiresa long-term perspective. As possible reorganization ofoperational structures, such as effective quality assessmentof microscopy and digitalization of the information sys-tem, would not take place before the end of an evaluationproject, further monitoring is indicated.

The fight against malaria is highly dependent on the par-ticipation of communities. It has long been known that

Active Case Detection in Amazonas (1988 - 2007)Figure 9Active Case Detection in Amazonas (1988 - 2007). Slides taken by active case detection (blue, left vertical axis), cases detected by active case detection (red, right vertical axis).

2000

2500

3000

3500

4000

20000

25000

30000

35000

40000

ted

by

Act

ive

Cas

e D

etec

tio

n

ctiv

e C

ase

Det

ecti

on

0

500

1000

1500

0

5000

10000

15000

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Mal

aria

Cas

es d

etec

t

Slid

es b

y A

c

Years

Page 11 of 14(page number not for citation purposes)

Page 12: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

community health workers (CHW) are key figures in theinterplay between the state's health system and communi-ties [43]. In this context, Venezuela looks back to a longtradition of training CHWs [44]. Interestingly, by now,two different approaches were observed within themalaria surveillance system of Amazonas. On one hand,salaried state health workers were trained as in manyother parts of the Venezuelan health system [45] and, onthe other hand, non-paid CHWs were trained followingconcepts of voluntary participation (see PAMAFRO) creat-ing ambiguities also seen in other parts of Venezuela [35].

A voluntary approach might be easily brokered to com-munities living in regions where the public health systemhas been commercialized or does not exist, as is the casein parts of Africa. However, it might be necessary to mod-ify this non-governmental approach in countries oppos-ing neoliberal reforms and showing the political will toestablish professionalized health care, through whichCHWs can make a living [46]. Probably, these conflictingapproaches explain the observed suboptimal performanceof voluntary CHWs in respect to slides taken. In order tostimulate voluntary participation it might be worth con-sidering avoiding the overlap of activities between profes-sional and voluntary CHWs. Slide taking was only one ofvarious activities by voluntary CHWs and a completeanalysis is underway.

Another clash of strategies was observed in respect to thedistribution of RDTs to CHWs and microscopists. The useof RDTs saves lives and reduces overtreatment in the Afri-can scenario, where microscopy is scarce. However, whena network of trained microscopists exists, the distributionof RDTs might cause microscopy to be neglected as wasseen in this RMA. In consequence, skills of trained micro-scopists and health workers could be lost rapidly. As lowlevel parasitaemia is common in the Venezuelan Amazonand reaches the detection limit of both methods, micros-copy and RDTs [18], the introduction of RDTs has to becarefully undertaken and monitored. In this scenario itmay be preferable to improve microscopy rather thanreplace it.

In conclusion, the example of the Venezuelan Amazonshows that - unlike many African countries, where impov-erishment and privatization have created different condi-tions for health interventions - the control of malaria ispossible, at least to a certain degree, and in the case thathealth facilities are working satisfactorily. The significantdecrease in slides taken in the years 2002 and 2003 (Fig-ure 2) shows how sensitively health parameters react tosocio-political circumstances: these years were charater-ized as the years of a failed coup d'etat and of the eco-nomic paralysis of the country. Other public health

Passive Case Detection in Amazonas (1988 - 2007)Figure 10Passive Case Detection in Amazonas (1988 - 2007). Slides taken by passive case detection (blue, left vertical axis), cases detected by passive case detection (red, right vertical axis).

5000

6000

7000

8000

9000

10000

50000

60000

70000

80000

90000

100000

ed b

y P

assi

ve C

ase

Det

ecti

on

ssiv

e C

ase

Det

ecti

on

0

1000

2000

3000

4000

0

10000

20000

30000

40000

19881989 1990 19911992 19931994 1995 19961997 1998 19992000 20012002 2003 20042005 2006 2007

Mal

aria

Cas

es d

etec

te

Slid

es b

y P

as

Years

Page 12 of 14(page number not for citation purposes)

Page 13: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

parameters in Venezuela, for example child mortality, alsoexperienced a clear negative impact in this period [47].

This RMA of Amazonas showed that, despite variousshortcomings and logistic problems because of limitedinfrastructure, the malaria surveillance system in the Ven-ezuelan Amazon provides a solid foundation for furtherinterventions. The estimation whether eradication orelimination of malaria will be possible in the VenezuelanAmazon or may, rather, be a distant prospect, was outsidethe scope of this RMA.

AbbreviationsACT: Artemisinin-based Combination Therapy; API:Annual Parasite Index; CHW: Community Health Worker;CML: Central Malaria Laboratory in Puerto Ayacucho;ICD: International Classification of Diseases and RelatedHealth Problems; ITN: Insecticide Treated Bed Nets; NP:Notification Point; PAMAFRO: Plan Andino de Malaria enAreas Fronterizas (Andean Malaria Plan in Frontier Areas);RDT: Rapid Diagnostic Test; RMA: Rapid MalariaAppraisal; RUMA: Rapid Urban Malaria Appraisal; SPR:Slide Positivity Rate.

Competing interestsThe authors declare that they have no competing interests.

Authors' contributionsWGM conceived the study, coordinated and conductedthe field work, analysed and interpreted data, and draftedand revised the manuscript. AMG participated in the con-ception of the work and coordinated the field work. SVMparticipated in the conception of the work, interpreted thedata, and revised it critically at all stages. JG was the keylocal contact person; he supervised the field activities. JACwas the administrator of the documentation system andparticipated in the cleaning of data sets. BGM contributedto the statistical analysis and revised the paper. MM facil-itated the overall collaboration and participated in theconception of the work. All authors read and approved thefinal manuscript.

Additional material

AcknowledgementsThe RMA and technical assistance was financially supported by UNICEF. We would like to thank Dr. Miguel Hernandez, all inspectors, micro-scopists, visitors, fumigators, drivers and staff of the malaria surveillance system of Amazonas who contributed to the success of the work. We thank John Williams for the mother-tongue revision of the manuscript.

References1. Alais ML: La población indígena de Venezuela según los cen-

sos nacionales. Encuentro nacional de demógrafos y estudiosos de lapoblación 2004.

2. Gabaldon A, Berti L: The First large area in the tropical zone toreport malaria eradication: north-central Venezuela. Am JTrop Med Hyg 1954, 3:793-807.

3. Hay SI, Smith DL, Snow RW: Measuring malaria endemicityfrom intense to interrupted transmission. Lancet Infect Dis2008, 8:369-378.

4. Sanchez E, Perez HA, Martinez C: Malaria in the Amazon. Prev-alence of Plasmodium falciparum antibodies in Amerindiansinhabiting the Venezuelan Amazon. Ann Trop Med Parasitol1990, 84:307-312.

5. Perez MS: Anemia and malaria in a Yanomami Amerindianpopulation from the southern Venezuelan Amazon. Am J TropMed Hyg 1998, 59:998-1001.

6. Freeman J, Laserson KF, Petralanda I, Spielman A: Effect of chemo-therapy on malaria transmission among Yanomami Amerin-dians: simulated consequences of placebo treatment. Am JTrop Med Hyg 1999, 60:774-780.

7. Laserson KF, Wypij D, Petralanda I, Spielman A, Maguire JH: Differ-ential perpetuation of malaria species among AmazonianYanomami Amerindians. Am J Trop Med Hyg 1999, 60:767-773.

8. Laserson KF, Petralanda I, Almera R, Barker RH Jr, Spielman A,Maguire JH, Wirth DF: Genetic characterization of an epidemicof Plasmodium falciparum malaria among Yanomami Amer-indians. J Infect Dis 1999, 180:2081-2085.

9. Torres R Jr, Magris M, Villegas L, Torres VM, Dominguez G: Spurcell anaemia and acute haemolysis in patients with hyperre-active malarious splenomegaly. Experience in an isolatedYanomamo population of Venezuela. Acta Trop 2000,77:257-262.

10. Tami A, Grundmann H, Sutherland C, McBride JS, Cavanagh DR,Campos E, Snounou G, Barnabé C, Tibayrenc M, Warhurst DC:Restricted genetic and antigenic diversity of Plasmodium fal-ciparum under mesoendemic transmission in the Venezue-lan Amazon. Parasitology 2002, 124:569-581.

11. Tami A, Ord R, Targett GA, Sutherland CJ: Sympatric Plasmodiumfalciparum isolates from Venezuela have structured var generepertoires. Malar J 2003, 2:7.

12. Marcano TJ, Morgado A, Tosta CE, Coura JR: Cross-sectionalstudy defines difference in malaria morbidity in twoYanomami communities on Amazonian boundary betweenBrazil and Venezuela. Mem Inst Oswaldo Cruz 2004, 99:369-376.

13. Ord R, Polley S, Tami A, Sutherland CJ: High sequence diversityand evidence of balancing selection in the Pvmsp3alpha geneof Plasmodium vivax in the Venezuelan Amazon. Mol BiochemParasitol 2005, 144:86-93.

14. Magris M, Rubio-Palis Y, Alexander N, Ruiz B, Galvan N, Frias D,Blanco M, Lines J: Community-randomized trial of lambdacy-halothrin-treated hammock nets for malaria control inYanomami communities in the Amazon region of Vene-zuela. Trop Med Int Health 2007, 12:392-403.

15. Rodulfo H, de DM, Mora R, Gonzalez L, Contreras CE: Comparisonof the diagnosis of malaria by microscopy, immunochroma-tography and PCR in endemic areas of Venezuela. Braz J MedBiol Res 2007, 40:535-543.

16. Rodulfo H, de Donato M, Quijada I, Peña A: High prevalence ofmalaria infection in Amazonas State, Venezuela. Rev Inst MedTrop Sao Paulo 2007, 49:79-85.

17. Magris M, Rubio-Palis Y, Menares C, Villegas L: Vector bionomicsand malaria transmission in the Upper Orinoco River,Southern Venezuela. Mem Inst Oswaldo Cruz 2007, 102:303-311.

Additional file 1Questionnaire. The questions about malaria knowledge were part of a larger questionnaire evaluating operational and organisational aspects and problems of the malaria surveillance system in Amazonas. They were translated from Spanish and are not in the original design.Click here for file[http://www.biomedcentral.com/content/supplementary/1475-2875-8-291-S1.PDF]

Page 13 of 14(page number not for citation purposes)

Page 14: A rapid malaria appraisal in the Venezuelan Amazon

Malaria Journal 2009, 8:291 http://www.malariajournal.com/content/8/1/291

Publish with BioMed Central and every scientist can read your work free of charge

"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."

Sir Paul Nurse, Cancer Research UK

Your research papers will be:

available free of charge to the entire biomedical community

peer reviewed and published immediately upon acceptance

cited in PubMed and archived on PubMed Central

yours — you keep the copyright

Submit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.asp

BioMedcentral

18. Metzger WG, Vivas-Martinez S, Rodriguez I, Goncalves J, Bongard E,Fanello CI, Vivas L, Magris M: Malaria diagnosis under field con-ditions in the Venezuelan Amazon. Trans R Soc Trop Med Hyg2008, 102(1):20-4.

19. Vivas L, O'Dea KP, Noya O, Pabon R, Magris M, Botto C, Holder AA,Brown KN: Hyperreactive malarial splenomegaly is associ-ated with low levels of antibodies against red blood cell andPlasmodium falciparum derived glycolipids in YanomamiAmerindians from Venezuela. Acta Trop 2008, 105:207-214.

20. Grenfell P, Fanello CI, Magris M, Goncalves J, Metzger WG, Vivas-Martinez S, Curtis C, Vivas L: Anaemia and malaria inYanomami communities with differing access to healthcare.Trans R Soc Trop Med Hyg 2008, 102:645-652.

21. Garcia-Casal MN, Leets I, Bracho C, Hidalgo M, Bastidas G, GomezA, Peña A, Pérez H: Prevalence of anemia and deficiencies ofiron, folic acid and vitamin B12 in an indigenous communityfrom the Venezuelan Amazon with a high incidence ofmalaria. Arch Latinoam Nutr 2008, 58:12-18.

22. Hay SI, Guerra CA, Gething PW, Patil AP, Tatem AJ, Noor AM,Kabaria CW, Manh BH, Elyazar IR, Brooker S, Smith DL, Moyeed RA,Snow RW: A world malaria map: Plasmodium falciparum ende-micity in 2007. PLoS Med 2009, 6:3 [http://map.nsms.ox.ac.uk/media/maps/pdf/limits/VEN_limits.pdf].

23. República Bolivariana de Venezuela Misión Médica Cubana: Informa-ción Estadística del 22 al 28 de Febrero de 2009. InformeFebrero, Caracas 2009.

24. Villalobos N: Analysis of the adjustment of capacity and physical conditionsof the Central Hospital of Puerto Ayacucho, Amazonas State, Venezuela in2007. Thesis Public Health. Hospital José Gregorio Hernández PuertoAyacucho; 2009.

25. WHO Regional Office for Europe: Gothenburg Consensus Paper: HealthImpact Assessment, Main Concepts and Suggested Approach EuropeanCentre for Health Policy, Brussels; 1999.

26. WHO: Monitoring and Evaluation Tool Kit. HIV/AIDS,Tuberculosis and Malaria. Geneva Second edition. 2009.

27. The R Development Core Team: R: A language and environmentfor statistical computing. 2009 [http://www.R-project.org]. RFoundation for Statistical Computing, Vienna

28. Ministerio de Poder Popular para la Salud: Anuario. Caracas 1998.29. Central Intelligence Agency (CIA): The World Fact Book. Washing-

ton; 2008. 30. Metzger WG: Informe Final de la Asistencia Técnica para la

Evaluación del Sistema de Vigilancia Epidemiológica e laMalaria en el Estado Amazonas. Caracas/Puerto Ayacucho;2008.

31. Plan Andino de Malaria en Areas Fronterizas (PAMAFRO): Present-ación del Proyecto "Control de la Malaria en las ZonasFronterizas de la Región Andina: Un Enfoque Comunitario".XVI Consejo Presidencial Andino. Lima 2005.

32. Barrera R, Grillet ME, Rangel Y, Berti J, Ache A: Temporal and spa-tial patterns of malaria reinfection in northeastern Vene-zuela. Am J Trop Med Hyg 1999, 61:784-790.

33. Zimmermann RH: Malaria in Sucre State, Venezuela. Cad SaúdePública 2000, 16:1127-1131.

34. Cáceres JL: Estado Sucre: El Éxito Antimalárico de Venezuelaen el Año 2003. Bóletin de Malariología y Salud Ambiental 2004,44:51-55.

35. Gómez I, Alarcón L, Velásquez Y: Cambios en la gestión del serv-icio antimalárico en el estado Sucre. Revista Venezolana de Ger-encia 2006, 11:277-303.

36. Greenwood BM: Control to elimination: implications formalaria research. Trends Parasitol 2008, 24:449-454.

37. Wang SJ, Lengeler C, Smith TA, Vounatsou P, Cisse G, Diallo DA,Akogbeto M, Mtasiwa D, Teklehaimanot A, Tanner M: Rapid urbanmalaria appraisal (RUMA) in sub-Saharan Africa. Malar J2005, 4:40.

38. Matthys B, Sherkanov T, Karimov SS, Khabirov Z, Mostowlansky T,Utzinger J, Wyss K: History of malaria control in Tajikistan andrapid malaria appraisal in an agro-ecological setting. Malar J2008, 7:217.

39. Wang SJ, Lengeler C, Smith TA, Vounatsou P, Diadie DA, Pritroipa X,Convelbo N, Kientga M, Tanner M: Rapid urban malaria appraisal(RUMA) I: epidemiology of urban malaria in Ouagadougou.Malar J 2005, 4:43.

40. Wang SJ, Lengeler C, Mtasiwa D, Mshana T, Manane L, Maro G, Tan-ner M: Rapid Urban Malaria Appraisal (RUMA) II: epidemiol-

ogy of urban malaria in Dar es Salaam (Tanzania). Malar J2006, 5:28.

41. Wang SJ, Lengeler C, Smith TA, Vounatsou P, Cisse G, Tanner M:Rapid Urban Malaria Appraisal (RUMA) III: epidemiology ofurban malaria in the municipality of Yopougon (Abidjan).Malar J 2006, 5:29.

42. Wang SJ, Lengeler C, Smith TA, Vounatsou P, Akogbeto M, TannerM: Rapid Urban Malaria Appraisal (RUMA) IV: epidemiologyof urban malaria in Cotonou (Benin). Malar J 2006, 5:45.

43. Bender DE, Pitkin K: Bridging the gap: the village health workeras the cornerstone of the primary health care model. Soc SciMed 1987, 24:515-528.

44. Yates AS: The Venezuelan Medicina Simplificada program.Public Health Rep 1975, 90:247-253.

45. Armada F, Muntaner C, Chung H, Williams-Brennan L, Benach J: Bar-rio Adentro and the reduction of health inequalities in Ven-ezuela: an appraisal of the first years. Int J Health Serv 2009,39:161-187.

46. PAHO/WHO: Mission Barrio Adentro: The Right to Health andSocial Inclusion in Venezuela. Regional Office of the World HealthOrganization. Caracas 2006.

47. Alvarado CH, Martínez M, Vivas-Martínez , Gutiérrez NJ, MetzgerWG: Social hange and health policy in Venezuela. Social Med-icine 2008, 2:95-109.

Page 14 of 14(page number not for citation purposes)